EUVIMEDEUVIMED entdecken
Crossref

Robust moment identification for nonlinear PDEs via a neural ODE approach

Shaoxuan Chen iD, Su Yang iD, Panayotis G Kevrekidis iD, Wei Zhu iD

Chaos: An Interdisciplinary Journal of Nonlinear Science · 01.12.2025 · 35 (12)

DOI: 10.1063/5.0288309 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Chaos: An Interdisciplinary Journal of Nonlinear Science, online erschienen am 19.12.2025. Verfasst von Shaoxuan Chen und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Mathematics and Statistics, University of Massachusetts Amherst, Amherst, Massachusetts 01003-4515, USA. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 56 Literaturangaben. Gefördert von: Division of Mathematical Sciences, Air Force Office of Scientific Research, Division of Physics.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We propose a data-driven framework for learning reduced-order moment dynamics from partial differential equation (PDE)-governed systems using neural Ordinary Differential Equations (neural ODEs). In contrast to derivative-based methods, such as SINDy (Sparse Identification of Nonlinear Dynamics), which necessitate densely sampled data and are sensitive to noise, our approach based on neural ODEs directly models moment trajectories, enabling robust learning from sparse and potentially irregular time series. Using as an application platform the nonlinear Schrödinger equation, the framework accurately recovers governing moment dynamics when closure is available, even from limited, irregular, and noisy observations. For systems without analytical closure, we introduce a data-driven coordinate transformation strategy based on Stiefel manifold optimization, enabling the discovery of low-dimensional representations in which the moment dynamics become closed, facilitating interpretable and reliable modeling. We also explore cases where a closure model is not known, such as a Fisher–Kolmogorov–Petrovskii–Piskounov reaction–diffusion system. Here, we demonstrate that neural ODEs can still effectively approximate the unclosed moment dynamics and achieve superior extrapolation accuracy compared to physical-expert-derived ODE models. This advantage remains robust even under sparse and irregular sampling, highlighting the method’s robustness in data-limited settings. Our results highlight the neural ODE framework as a powerful and flexible tool for learning interpretable, low-dimensional moment dynamics in complex PDE-governed systems.

Autor:innen und Institutionen

  • Department of Mathematics and Statistics, University of Massachusetts Amherst, Amherst, Massachusetts 01003-4515, USA.
  • Department of Physics, University of Massachusetts Amherst, Massachusetts 01003-9337, USA.
  • Theoretical Sciences Visiting Program, Okinawa Institute of Science and Technology Graduate University, Onna 904-0495, Japan.
  • School of Mathematics, Georgia Institute of Technology, Atlanta, Georgia 30332, USA.

ORCID-Profile: Shaoxuan Chen, Su Yang, Panayotis G Kevrekidis, Wei Zhu

Förderung

  • Division of Mathematical Sciences
  • Air Force Office of Scientific Research
  • Division of Physics
  • Simons Foundation

Zitieren

APA

Chen, S., Yang, S., Kevrekidis, P. G., & Zhu, W. (2025). Robust moment identification for nonlinear PDEs via a neural ODE approach. Chaos: An Interdisciplinary Journal of Nonlinear Science, 35(12). https://doi.org/10.1063/5.0288309

Vancouver

Chen S, Yang S, Kevrekidis PG, Zhu W. Robust moment identification for nonlinear PDEs via a neural ODE approach. Chaos: An Interdisciplinary Journal of Nonlinear Science. 2025;35(12). doi:10.1063/5.0288309

BibTeX
@article{Chen2025,
  author = {Chen, Shaoxuan and Yang, Su and Kevrekidis, Panayotis G and Zhu, Wei},
  title = {Robust moment identification for nonlinear PDEs via a neural ODE approach},
  journal = {Chaos: An Interdisciplinary Journal of Nonlinear Science},
  year = {2025},
  volume = {35},
  number = {12},
  doi = {10.1063/5.0288309}
}
Kostenlos starten →